Microfluidic nanostrain sensor for the detection, enumeration, and characterization of circulating tumor cells in patient’s whole blood
Introduction
The enumeration and characterization of circulating tumor cells (CTCs) can provide clinical guidance in terms of early detection, prognosis, and treatment guidance. Most of the current CTC enrichment technologies first need to isolate CTCs from other blood cells and then perform analysis such as enumeration and characterization. This strategy can induce biases in the analysis results due to the isolation methodology. For example, an affinity isolation method will only capture CTCs that express specific proteins whereas CTCs smaller than a certain size will be lost when employing a physical isolation method. In this study, we proposed a microfluidic nanostrain sensor that bypasses the isolation step and directly detects, enumerates, and characterizes CTCs in patients’ whole blood on-the-fly.
Methods
The microfluidic nanostrain sensor is comprised of three parts: a glass microfluidic with a straight constriction channel for cells to flow through, a nm thick wrinkle-free metallic thin film in polymer (WiMTiP) sensor where the microfluidic channel is bonded to, and a digital holography interferometry (DHI). As CTCs flow through the constriction, they will deform the entire WiMTiP sensor including the encased nanometer metallic (e.g. Al) thin film. The deformation of the embedded thin film, which acts as a flexible mirror, is read out non-intrusively by a holographic interferometry. The obtained interferogram is further processed digitally to obtain the 3D deformation profiles caused by passing CTCs. A two-phase flow with fluid-structure interaction model in COMSOL is used to simulate the detection mechanisms and operation parameters of the sensor. Two cancer cell lines, PC3 and LnCap, are used to validate the simulation results. The detection and enumeration of CTCs are done by analyzing the cross-correlation map statistics whereas the characterization of CTCs is based on the surface deformation profiles.
Results
The proposed microfluidic nanostrain sensor has a resolution of 2.3 nm with a 1.5 mm × 1.5 mm field of view. The simulation results reveal CTCs will be deformed when passing through the constriction and this will form a plug flow in the channel, resulting in a pressure build-up in the trailing edge of the cell and a pressure reduction in the leading edge of it. Both increase and reduction leave distinctive deformation imprints on the sensor. More importantly, the sensor deformation is dependent on CTC cell size and its membrane viscoelastic properties, both of which CTCs have different characteristics when compared to that of blood cells. Experimental confirmation of the simulation results will be reported using two prostate cancer cell lines PC3 and LnCap.
Conclusion
The microfluidic nanostrain sensor is a promising tool for the detection, enumeration, and characterization of CTCs on-the-fly without introducing biases. Genomic and functional analysis can be performed on the CTCs since there is no alteration to the cancer cells. Further optimization is being performed to increase the throughput and the performance of the sensor.